What's going on with these PSU measurements?

[Steve]

Hi all,

I'm just trying to measure the output of an Amiga 312503 PSU ( schematics: https://github.com/cammeltech/-006-comm ... -03_schema )

As I'm going to perform a capacitor upgrade, I just want to see the improvement on my scope. The default multi meter measurements look fine at 5v / 11.5v. I've hooked it up with a 5v and 12v load resistor.

The results look really screwy, could someone please tell me what I'm doing wrong. Am I using the scope properly? (I do press auto of course, and then just adjust the scales to suit, using 10x probe)

FYI the PSU does work in an Amiga...

NewFile2.jpg

NewFile1.jpg

NewFile0.jpg

1.jpg

[JezC]

Might be picking up the mains (50Hz) - if you use a much slower time base then you can check if it is the problem.

If so, you can usually filter out the AC component.

Also check where you are connecting the probe ground.

[Badwolf]

The fine oscillations could be the switching frequency. Looks around 30kHz. Could be a bit of aliasing too.

The slower ones could be the result of testing without load?

You want a much slower timebase really. Check you're using DC coupling. Check bandwidth filters. Check averaging is off.

BW

[Steve]

@JezC the probe ground is going to GND (not shield) in the PSU (Amiga PSUs have both gnd and shield, am I supposed to use one or the other, I've chosen gnd so far)

@Badwolf I've load resistors on 5v and 12v. In regards to: DC coupling. bandwidth filters. averaging - I will have to research what these mean and find out. (I am fairly new to scopes)

FYI I am trying to measure ripple/noise, just like Exxos did in his Atari PSU experiments. I am unsure as to what settings I should be using.

[exxos]

Vmax shows 4.88V in one image. You need to go into the measure menu and select Vmin as well.

It kinda looks like the voltages varying between like 4 and 5V :shrug: if it is doing that then the output capacitors have failed.

[Badwolf]

@Badwolf I've load resistors on 5v and 12v. In regards to: DC coupling. bandwidth filters. averaging - I will have to research what these mean and find out. (I am fairly new to scopes)

FYI I am trying to measure ripple/noise, just like Exxos did in his Atari PSU experiments. I am unsure as to what settings I should be using.

Try a 1ms timebase, the 1 V/div is fine, providing both the scope and your probes agree you're in 10x mode.

The bandwidth limit is in the channel configuration, IIRC. Press the CH1 button a couple of times. First it'll turn the channel off, then it'll turn it on with a menu -- I think, don't have my scope to hand. Turn that bandwidth limit ON (the proper high frequency stuff is pretty much irrelevant for ripple measurement, and you'll just pick up all the background sh*te going on). Averaging may be in there as well. You want that off, else you'll get *great* results. :)

BW

[rubber_jonnie]

Usually when I see weird results it's the probe GND that's the issue.

Make sure the leads gnd leads are good.

[Steve]

Usually when I see weird results it's the probe GND that's the issue.

Make sure the leads gnd leads are good.

Maybe I need to connect to shield... Amigas have gnd/shield seperate, all these measurements were connected to gnd.

[rubber_jonnie]

Usually when I see weird results it's the probe GND that's the issue.

Make sure the leads gnd leads are good.

Maybe I need to connect to shield... Amigas have gnd/shield seperate, all these measurements were connected to gnd.

Worth a try for sure, at least eliminate the simple stuff.

Just recapping my A1200 supply, but the dumbass that's me only ordered 2 of one type of cap when there's 3...

[Steve]

@rubber_jonnie are you maybe planning on recording before/after readings like me? I don't think anyone in the Amiga community has done that kind of thing yet.